Académique Documents
Professionnel Documents
Culture Documents
Article views: 19
Transient dynamic impact suppression of a Baja chassis using frontal and rear
shock absorbers
a
K. M. Goher , Kong Chenhuib, S. O. Fadlallahc, A. M. Al Shabibid and N. Z. Al Rawahid
a
Department of Informatics and Enabling Technologies, Lincoln University, New Zealand; bMechanical Engineering Department, University of
Shanghai for Science and Technology, Shanghai ; cMechanical Engineering Department, Auckland University of Technology, New Zealand;
d
Department of Mechanical and Industrial Engineering, College of Engineering, Sultan Qaboos University, Muscat, Oman
Finite element analysis has been widely applied to the simulation and analysis: for a Chevrolet Silverado pick-
simulation of vehicle crash. Several research studies up truck [25], for a Honda Accord DX Sedan [6], for
investigated how to develop a full vehicle model for both Toyota Yaris passenger car and Ford Explorer SUV
crashworthiness analysis by Finite Element Analysis [20], and for Ford Crown Victoria [18] to study the
(FEA) [3,6,11]. Borovinsek et al. [4] provided a brief effect on the overall crash safety of lightweight passenger
overview of different types of vehicle collisions. Crash cars.
FEMs are often the rst detailed models developed in This paper investigates the usability of springs, which
the design process of road cars [9]. Usually, the crash exhibit nonlinear force-deection characteristic, in the
models are rst validated by the corresponding impact area of mathematical modelling of vehicle crash. We
tests at the component level [18] and material plasticity present a method, which allows obtaining parameters of
parameters are the major focus. The discrepancies the spring-mass model basing on the full-scale experi-
between FEM and actual vehicle structure usually come mental data analysis for a Baja car chassis. Since vehicle
from, on the hardware side, missing parts, initial parts collision is a dynamic event, it involves such phenomena
penetration, geometric variation, welding characteristics, as rebound and energy dissipation. Three different
etc.; on the FEM side, material properties, element qual- spring unloading scenarios (elastic, plastic and elasto-
ity, concentrated mass distribution, etc. [6]. plastic) are covered and their suitability for vehicle colli-
Cosme et al. [8] performed specic case studies in the sion simulation is evaluated. Subsequently, we assess
design and analysis of heavy-duty frames. Karaoglu and which of those models ts the best to the real car's
Kuralay [17] performed stress analysis of a truck chassis behaviour not only in terms of kinematic responses but
by using FEM. Livemore Software Dynamic Analysis also in terms of energy distribution. This paper is organ-
(LS-DYNA) is an explicit code highly capable of solving ised in four different sections. Section 1 gives a short
high-speed impact problems that require small time description of the objective and the problem. In Section
steps, which is commonly used by researchers in vehicle 2, the mathematic model was chosen and the method of
modelling, analysis and crashworthiness evaluation [19]. impact force calculation was displayed along with all the
Argyris et al. [1,2] presented the theoretical background settings related to modelling and dynamic analysis pro-
for implicit nite element formulation and proposed a cedure. Results of the dynamic analysis are shown in
crash test analysis using simplied shell elements. The Section 3. Finally, Section 4 contains the conclusion of
rst published simulation of a vehicle frontal impact this project.
dates back to 1986, when Haug et al. [14] simulated a
frontal crash of a Volkswagen Polo. Jenefeldt and Thom-
2. Chassis modelling
son [15] investigated the methodology to match frontal
stiffness levels for vehicles in frontal impacts with differ- There are several software packages that are equipped to
ent mass ratios. As for Thomson et al. [27], their study handle the crash-testing of vehicles but one of the most
focused on car-to-car frontal crash compatibility. popular is from ANSYS Software Inc. called ANSYS
Workbench. With explicit dynamics application, auto-
motive companies and their suppliers can test car
1.2. Car modelling
designs without having to tool or experimentally test a
FEMs of vehicles and vehicle components have been prototype, thus saving time and expense [22]. But the
increasingly applied in preliminary design analysis, vehi- explicit dynamic solver does not support spring and
cle crashworthiness evaluation and component design. damper connection [12], and considering the shock
The rst successful lumped parameter model for the absorbers in our project, transient structural solver is
frontal crash of an automobile was developed by Kamal chosen to simulate the collision. Unlike in explicit solver,
[16]. Pawlus et al. [23] proposed a basic mathematical in transient structural solver, all solids have to be set as
model to represent a collision together with its analysis. exible, including chassis, bumper and wall. So the prob-
The rst collision simulation model was developed to lem is when the real collision happens, the wall is sup-
run on a digital computer in the mid-1970s [28]. Finite posed to be rigid. To solve this problem, the initial
element analysis was then introduced to calculate the velocity on the bumper is set instead of using the rigid
complicated deforming process in crashes [13]. Current wall to hit the chassis.
approach to car crash simulation is usually based on
nite element analysis with an emphasis on better per-
2.1. SOLIDWORKS modelling
formance and safety, and thus, reducing the risk of fatal
injuries during an accident [10]. Number of researchers The commonly followed steps for modelling a vehicle
have created high delity vehicle models for crash assembly can be dened as follows: dimension
INTERNATIONAL JOURNAL OF CRASHWORTHINESS 3
initial velocity is applied to the mass of the vehicle. As Since d vinitial t, substituting this in Equation (5)
shown in Figure 3(a,b), the vehicle moves forwards or leads to
backwards at a constant speed until it strikes the rigid
wall. As for the third scenario, combined frontal and rear 1 1
F mvinitial (6)
impacts, two vehicles with the same initial speed from 2 t
both sides collide with the middle vehicle. In this study, it
is assumed that the shock absorber in the collision side is With the information of the vehicle's total mass
capable of absorbing the energy during the impact. m, initial impact velocity vinitial and considering
that the vehicle comes to rest 0.1 seconds after the
impact t 0:1 seconds, the impact force Fcan be
2.4. Impact force evaluation calculated.
The impact force, for a perfectly inelastic collision, can The weights of all the components of the Baja
be estimated using the following equations [21]: vehicle are evaluated as shown in Table 2 and equiva-
lent weights are modelled in the form of solid rigid
1 2 1 2 blocks. Considering the vehicle's initial speed as 70
WNET mvfinal mvinitial (3)
2 2 km/hr (19.4 m/s), and with the total weight of all
WNET F d (4) components, the impact force (F) can be calculated
using Equation (6):
Equation (3) states that the variation in kinetic energy
is equal to the net work done WNET , and the work Table 2. Weight of components of the Baja vehicle [21].
needed to stop the car (Equation (4)) is equal to the Component Weight (kg)
impact forceFtimes the distanced. Combining these Equivalent mass of chassis (including body sheet and 54.4
two equations, and given the fact that the nal impact gussets)
Equivalent mass of engine (Briggs & Stratton 10 hp) 27.2
velocity vfinal is zero, leads to the following: Equivalent mass of tire assembly (includes suspension) 45.4
Equivalent mass of transmissiosn 27.2
Equivalent mass of steering, brakes and auxiliary 27.2
1 2 1
F mvinitial (5) Equivalent mass of driver 90.7
2 d Total mass 272.1
INTERNATIONAL JOURNAL OF CRASHWORTHINESS 5
In rear only collision, the highlighted columns of range of 80120 N/mm. Focusing on the located zone
the rear chassis (Figure 6(b)) are just behind the and with a step size of 5 N/mm (Figure 7(b)), the front
occupant seat. The deformation of these columns beam's maximum deformation drop occurred in the
may endanger the occupant's back if the seat cannot zone between k1 = 105 and 115 N/mm. Repeating the
absorb all the collision energy. previous step again by magnifying the targeted range
In frontal and rear collision, both highlighted fea- and considering a 1 N/mm increment in the stiffness
tures should be considered. values, it can be observed from Figure 7(c) that the mini-
mum deformation took place at k1 = 113.5 N/mm. For
this scenario, it can be concluded that the suitable stiff-
3. Results and discussion ness value for the front shock absorber is approximately
3.1. Frontal collision only k1 = 113.5 N/mm.
deformation. For multiple stiffness values for the rear stiffness values (Figure 8(b)), the maximum deformation
shock absorber in the range of k2 = 20200 N/mm drop occurred in the stiffness range of 106109.5 N/
(Figure 8(a)), the minimum deformation was located in mm. Again, by repeating the same procedure but with a
the zone between k2 = 102 and 112 N/mm. Magnifying step size of 0.5 N/mm (Figure 8(c)), the minimum defor-
the specied range with a 2 N/mm increment in the mation occurred at k2 = 108 N/mm. Therefore, for this
INTERNATIONAL JOURNAL OF CRASHWORTHINESS 9
Figure 9. Front beam and rear columns deformation under frontal and rear collision (k1 = k2 = k).
10 K. M. GOHER ET AL.
Figure 10. Front beam and rear columns deformation under frontal and rear collision (k1 6 k2).
INTERNATIONAL JOURNAL OF CRASHWORTHINESS 11
Figure 11. ANSYS Workbench results for k1 = 160 N/mm, k2 = 145 N/mm.
scenario, the best stiffness value for the rear shock assumes that both front and rear shock absorbers have
absorber is around k2 = 108 N/mm. the same stiffness value (k1 = k2 = k). As for the other
case, it considers that the stiffness is different for the two
shock absorbers (k1 6 k2).
3.3. Frontal and rear collision
For this scenario, where the Baja vehicle collides with 3.3.1. Case 1 (k1 = k2 = k)
two vehicles with the same initial speed from both front For this case, the stiffness of both front and rear shock
and rear sides, two cases were investigated. The rst case absorbers is considered to be the same (k1 = k2 = k).
12 K. M. GOHER ET AL.
Figure 9(a) shows the front beam and rear columns beam and rear columns were achieved as 0.14387 and
deformation under frontal and rear impact for multiple 0.058255 mm respectively.
stiffness values for both front and rear shock absorbers
varying between k = 60 and 180 N/mm. Based on the
previous results, the maximum deformation drop 4. Conclusion and future work
occurred between k = 100 and 140 N/mm. Concentrat- The overall objective of the work was to seek the suitable
ing on the specied zone and setting a 5 N/mm incre- parameters of stiffness and damping coefcient in shock
ment in the stiffness values (Figure 9(b)), the front beam absorbers employed to a Baja vehicle chassis. By utilising
and rear columns maximum deformation drop was ANSYS Workbench software package, simulations were
located in the stiffness range of 125132 N/mm. Focus- performed. Due to the difculty in transferring the
ing on the located segment and with a step size of 1 N/ actual Baja car model into ANSYS Workbench, a simpli-
mm (Figure 9(c)), the minimum deformation in both ed model of the test vehicle was created and simula-
the front beam and the rear columns occurred when the tions were carried out for three scenarios: frontal impact
stiffness value was at k = 131 N/mm. Based on the only, rear impact only, and both frontal and rear
dened conditions, the suitable stiffness value for both impacts. According to simulation results, and based on
front and rear shock absorbers is 131 N/mm. the obtained values of stiffness for both frontal and rear
shock absorbers, minimum deformation of the chassis
3.3.2. Case 2 (k1 6 k2) front beam and rear columns was achieved, and there-
The stiffness of both front and rear shock absorbers is fore, it can be concluded that the Baja car's occupant
considered to be different for this case (k1 6 k2). It has would not be harmed by the deformation of the chassis
been observed in the previous scenarios, frontal collision crucial parts. Future considerations of this work will
only and rear collision only, that the stiffness of the front consider increasing the model's complexity, investigating
shock absorber is always greater than that of the rear the effect of varying the impact speed on the vehicle's
shock absorber (k1 > k2). Therefore, and based on this chassis, and validating the achieved simulation results
observation, the rear shock absorber's stiffness (k2) was with experimental ones.
congured rst and the obtained value was employed to
seek the front shock absorber's stiffness (k1). Figure 10
(a) demonstrates the front beam and rear columns Disclosure statement
deformation under frontal and rear collision for multiple No potential conict of interest was reported by the authors.
stiffness values for the rear shock absorber varying
between k2 = 40 and 200 N/mm. The maximum defor-
mation drop occurred in the range of k2 = 130150 N/ ORCID
mm. Focusing on the selected area and setting a 5 N/
K. M. Goher http://orcid.org/0000-0003-2332-3876
mm increment in the stiffness values (Figure 10(b)), the
minimum deformation in both front beam and rear col-
umns took place at k2 = 145 N/mm. Using the obtained References
value of k2, Figure 10(c) shows the front beam and rear
columns deformation under frontal and rear collision [1] J. Argyris, H. Balmer, and I.St. Doltsinis, Some thoughts
on shell modelling for crash analysis, Comput. Method
for multiple stiffness values for the front shock absorber
Appl. Mech. Eng. 71 (1988), pp. 341365.
changing between k1 = 40 and 200 N/mm. As can be [2] J. Argyris, H.A. Balmer, J.St. Doltsinis, and A. Kruz,
seen, the range where the deformation reaches its mini- Computer simulation of crash phenomena, Int. J. Numer.
mum is within 150180 N/mm. Zooming in the speci- Methods Eng. 22 (1986), pp. 497519.
ed range (Figure 10(d)), the maximum deformation [3] A.O. Atahan, Crashworthiness analysis of a bridge rail-to-
drop was located at k1 = 160 N/mm. Therefore, and guardrail transition, Int. J. Crashworthiness. 21 (2016),
pp. 112.
according to the obtained results, the suitable stiffness [4] M. Borovinsek, M. Vesenjak, M. Ulbin, and Z. Ren, Sim-
values for both front and rear shock absorbers are k1 = ulation of crash tests for high containment levels of road
160 N/mm and k2 = 145 N/mm. safety barriers, Eng. Fail. Anal. 14(8) (2007), pp. 1711
Figure 11 represents ANSYS Workbench simulation 1718.
results for both front beam and rear columns in terms of [5] P. Cheng, M. Sens, J. Wiechel, and D. Guenther, An over-
view of the evolution of computer assisted motor vehicle
directional deformation, equivalent elastic strain and
accident reconstruction, SAE International, Warrendale,
equivalent stress after applying the obtained stiffness val- PA, 1987.
ues of both front and rear shock absorbers. Based on the [6] Z.Q. Cheng, J.G. Thacker, W.D. Pilkey, W.T. Hollowell,
results, minimum deformations of the Baja chassis front S.W. Reagan, and E.M. Sieveka, Experiences in reverse
INTERNATIONAL JOURNAL OF CRASHWORTHINESS 13
engineering of a nite element automobile crash model, Papers, Paper number 2000-01-0627, SAE International,
Finite Elem. Anal. Des. 37 (2001), pp. 843860. Warrendale, PA, USA, 2000.
[7] Y.B. Cho, C.H. Bae, M.W. Suh, and H.C. Sin, Maximisa- [19] L. Mei and C.A. Thole, Data analysis for parallel car-
tion of crash energy absorption by crash trigger for vehicle crash simulation results and model optimization, Simul.
front frame using the homogenisation method, Int. J. Model. Pract. Theory 16 (2008), pp. 329337.
Vehicle Des. 46 (2008), pp. 2350. [20] M. Mongiardini, R.H. Grzebieta, G.A. Mattos, and M.R.
[8] C. Cosme, A. Ghasemi, and J. Gandevia, Application of Bambach, Computer modelling of vehicle rollover crash
computer aided engineering in the design of heavy-duty tests conducted with the UNSW Jordan Rollover System,
truck frames, SAE Transactions 1999-01-3760, SAE Int. J. Crashworthiness 21 (2015), pp. 118.
International, Warrendale, PA, USA, 1999. [21] N. Nagurbabu, Computational analysis for improved
[9] S. Donders, Y. Takahashi, R. Hadjit, T. VanLangenhove, design of an SAE BAJA frame structure, Master thesis,
M. Brughmans, B. Van Genechten, and W. Desmet, A University of Nevada, 2010.
reduced beam and joint concept modelling approach to [22] M. Ohashi, Future trend of automobile and the high
optimize global vehicle body dynamics, Finite Elem. Anal. strength sheet steel, Tetsu-to-Hagane 68(9) (1982), pp.
Des. 45 (2009), pp. 439455. 11361146.
[10] G. Genta, Motor Vehicle Dynamics: Modelling and Simu- [23] W. Pawlus, J.E. Nielsen, H.R. Karimi, and K.G. Robber-
lation, World Scientic, London, 1997. smyr, Mathematical Modeling and Analysis of a Vehicle
[11] K.T. Gursel and S.N. Nane, Non-linear nite element Crash, Proceedings of the 4th European Computing
analyses of automobiles and their elements in crashes, Int. Conference, In M. Grigoriu (Ed.),WSEAS Press, Buchar-
J. Crashworthiness 15 (2010), pp. 667692. est, Romania, 2010, pp. 194199.
[12] J.O. Hallquist, LS-DYNA Keyword User's Manual, Liver- [24] M. Peden, R. Scureld, D. Sleet, D. Mohan, A.A. Jyder, E.
more Software Technology Corporation, Livemore, Cali- Jarawan, and C. Mathers, World Report on Road Trafc
fornia, USA, (2007) pp. 970. Injury Prevention, WHO, Geneva, 2004. Available
[13] J. Happian-Smith, An Introduction to Modern Vehicle athttp://apps.who.int/iris/bitstream/10665/42871/1/
Design, Reed Educational and Professional Publishing 9241562609.pdf.
Ltd., Oxford, 2002. [25] M. Soltani, A. Topa, M.R. Karim, and N.R. Sulong,
[14] E. Haug, T. Scharnhorst, and P. DuBois, FEM-crash Crashworthiness of G4 (2W) guardrail system: a nite ele-
Berechnung eines Fahrzeug frontalaufpralls, VDI- ment parametric study, Int. J. Crashworthiness, 22
Tagung: Berechnung im Automobilbau (VDI-Tagung: (2016), pp. 121.
Calcculation in Automotive Engineering) 613 (1986), pp. [26] P. Spethmann, S.H. Thomke, and C. Herstatt, The impact
479505. of crash simulation on productivity and problem-solving
[15] F. Jenefeldt and R. Thomson, A methodology to assess in automotive R&D, Working Paper 43, Technologie-
frontal stiffness to improve crash compatibility, Int. J. Und Innovations Management, Technische Universitat
Crashworthiness 9 (2004), pp. 475482. Hamburg- Harburg, Hamburg, 2006.
[16] M.M. Kamal, Analysis and simulation of vehicle to bar- [27] R. Thomson, M. Edwards, T. Martin, C. van der Zweep,
rier impact, SAE Technical Papers, Tech. Paper 700414, R. Damm, and G. della Valle, Carcar crash compatibility:
SAE Internationa, , Warrendale, PA, USA, 1970. development of crash test procedures in the VC-compat
[17] C. Karaoglu and N.S. Kuralay, Stress analysis of a truck project, Int. J. Crashworthiness 12 (2007), pp. 137151.
chassis with riveted joints, Finite Elem. Anal. Des. 38 [28] R. York and T.R. Day, The DyMESH method for three-
(2002), pp. 11151130. dimensional multi-vehicle collision simulation, SAE Tech-
[18] S.W. Kirkpatrick, Development and validation of high nical Papers, Paper number: 1999-01-0104, SAE Inter-
delity vehicle crash simulation models, SAE Technical national, Warrendale, PA, USA, 1999.